ILTA 2010 Strategic Unity. The Guru s Guide for Helpdesk and Deskside Support

Transcription

1 ILTA 2010 Strategic Unity The Guru s Guide for Helpdesk and Deskside Support 8/26/2010

2 Introduction You can only manage what you measure. Measuring service levels, ticket volumes, and user satisfaction is only valuable if it can be measured over time, compared and analyzed to other metrics, and acted upon. Gaining access to the needed data is critical to any managed process. Having and utilizing the necessary tools are imperative to providing effective and efficient helpdesk and deskside support. This report was compiled by Intelliteach, a legal-specific helpdesk outsourcing company. The data represents actual law firm results across a variety of law firm sizes, locations, and hardware and software configurations for the period of January 1 June 30, Items of note: An Automatic Call Distribution (ACD) system receiving, distributing and reporting the results of all phone and support requests is in place. A ticketing system automatically recording each customer interaction and soliciting end-user satisfaction is in place (IQTrack). Staffing analyses are based on Erlang-C traffic models Assumed Service Levels are as follows: >= 90% Telephone Live Rate within 20 seconds <= 5 minute average wait all others calls <= 15 minute average response time all s >= 85% first call resolution >= 90% excellent or good end-user survey Any changes to the above systems or assumptions will alter the corresponding data

3 Top Ticket Categories Resolved Helpdesk (red) compared to the rest of IT (blue) Microsoft Outlook 2003 Microsoft Word 2003 Document Management System Printer Administrative Network Document Conversion Desktop Exchange Server Remote Access Software Equipment Request Microsoft Internet Explorer Installation Microsoft Windows XP Telephone Microsoft Excel 2003 Laptop Document Management System PDA Remote Access Software Non Core Applications Network Password Reset Microsoft Word 2007 Microsoft Outlook 2003 Document Comparison Microsoft Outlook 2007 Timekeeping Microsoft PowerPoint 2003 Digital Voice Recording Records File Request

4 Top Ticket Categories Resolved (continued) Items of note: Over 46% of all service desk tickets resolved in the first half of 2010 are specific to various versions of Microsoft Office. Comparatively, the number of tickets resolved by the helpdesk is much higher than the rest of IT. How can even more of these be moved to the desk? 11% of tickets opened and resolved by IT pertain to common printer issues. This is historically the biggest resource drain of any IT department and the most strategic area to outsource.

5 8am 9am 10am 11am 12pm 1pm 2pm 3pm 4pm 5pm 8am 9am 10am 11am 12pm 1pm 2pm 3pm 4pm 5pm The Guru's Guide for Helpdesk and Deskside Support How Conversions Impact Staffing, Quality of Service, or Both Conversions cause a significant increase in volume and add additional strain to existing resources and systems. Most firms prudently staff to maintenance levels not project levels and require additional support to properly service the user community during upgrades. Using recent data from law firm helpdesks (average firm size 700 users), volumes increased by 42.5% during firm-wide software upgrades. The impact to staffing is significant. The difficulties a firm faces in recruiting and training a team familiar with the new software while maintaining existing support for the old software is only magnified by the recent cost cutting measures most firms have instituted. Pre-Conversion Volume with Staffing to maintain Service Levels Pre- Conversion Volume Staffing Conversion Volume with Staffing to maintain Service Levels Conversion Volume Staffing

6 8am 9am 10am 11am 12pm 1pm 2pm 3pm 4pm 5pm 8am 9am 10am 11am 12pm 1pm 2pm 3pm 4pm 5pm The Guru's Guide for Helpdesk and Deskside Support How Conversions Directly Impact Staffing, Quality of Service, or Both (continued) Another approach is simply to make do with what you already have. If a firm assumes they can absorb the volume increase, not only will the IT staff suffer, but response time to end-users and possibly the quality of support will be negatively impacted. Using the same data above, we modeled Service Levels and how those compare without appropriate staffing. Pre-Conversion Volume with Pre-Conversion Staffing Pre-Conversion Calls Answered within 20s 96% 92% 94% 92% 93% 93% 93% 93% 95% 94% Average Delay (seconds) Max Time in Queue (seconds) Percent Queued 6% 10% 8% 10% 9% 8% 8% 9% 7% 8% Conversion Volume with Pre-Conversion Staffing Conversion Calls Answered within 20s 81% 61% 65% 60% 67% 71% 70% 67% 76% 77% Average Delay (seconds) Max Time in Queue (seconds) Percent Queued 21% 43% 38% 43% 36% 33% 34% 36% 28% 26% The overall summary to this approach reduces the live rate by 25% and significantly increases user hold times, abandon rates, and maximum queue times all contributing to helpdesk frustration and a dissatisfied user experience. Measurement Pre-Conversion Conversion Difference Overall Live Rate 93% 68% -25% Average Time in Queue 0:00:10 0:01:21 0:01:11 Maximum Time in Queue 0:01:40 0:12:40 0:11:00

7 Tickets by Origin The Guru's Guide for Helpdesk and Deskside Support Tickets by Origin represent how the end-user contacted the helpdesk. Over 200,000 tickets were analyzed to determine the following data. Almost 70% of contacts were via the telephone while produced 28% of total contacts. 3% Other include in-person visits or phone and contacts made to other members of the IT staff. is the fastest growing Origin with most firms averaging less than 10% just five years ago. Other, 3.0% Voic , 0.3% , 28.2% Telephone, 68.5% Items of note: While continues to grow in usage, are users actually receiving better service? Data indicates that, on average, tickets created via have a lifespan 6 times longer than those originated by telephone. This is typically due to the user s inability to troubleshoot their situation or clearly describe the need they have in their original request, often requiring an additional exchange or callback.

8 Tickets and Requesters by Location Measuring the effectiveness of support across locations is an important gauge on how successful an IT organization is at reaching all of its users and determining if centralized procedures are paying dividends. A simple chart, by location, providing % of users, % of tickets, and % of unique users will provide invaluable information to that end. Location Total Users Total Tickets Unique Users New York 51% 67% 80% Chicago 23% 18% 48% Los Angeles 26% 15% 71% Graphing the same data will often provide clarity to any issue which in need of review. 80% 70% 60% 50% 40% 30% 20% Users Tickets Unique Users 10% 0% New York Chicago Los Angeles Items of note: While the New York office represents 51% of the total users in the firm, its users produce 67% of the total volume. Why is this? Could that location being going through an upgrade, have networking issues that haven t previously been identified or could they simply need more assistance? Why are Chicago s unique users much lower than other locations? Could it be that a local IT representative is fielding requests that normally should go to the help desk (and not being logged)?

9 Surveys Measuring user satisfaction may be the most important measurement of all. If users aren t happy with the service they receive they will find an alternative for support in the form of a co-worker or possibly discover the answer for themselves. Both of these alternatives create non-productive time while never addressing rootcause issues and/or sharing knowledge with others. Providing your end-user community the opportunity to rate each and every interaction with your helpdesk is critical in effective management of that function. Having ticket-specific reporting vs. annual or occasional surveys allow for action to be taken whether the issue might be specific to a particular analyst, a lack of teamwide training, training the entire firm might benefit from, or sometimes an adjustment to user expectations. Below are survey results which included over 200,000 tickets with a return rate of 17% (over 34,000 end-user survey responses). It is imperative that once survey information is collected that any negative items are addressed immediately. In our environment, poor surveys automatically reopen, the priority changes to emergency status, and the ticket is escalated to a member of management. Regardless of how you handle your workflow never allow a complaint to go unanswered. 100% 90% 80% 70% 60% 50% 40% 30% 20% Excellent Good Fair Needs Improvement 10% 0% Quality of Service Timeliness Friendliness Overall Experience

10 About the Author Lance Waagner is the President and CEO of Intelliteach. Prior to founding Intelliteach in 1998, Lance served as CIO for a 450-lawyer firm. He is an active vendor partner of the International Legal Technology Association (ILTA) and periodically writes and speaks on legal helpdesk and service delivery topics. He can be reached at lwaagner@intelliteach.com. Intelliteach specializes in legal-specific helpdesk outsourcing since 1998 and as such has collected massive amounts of ticketing data and helpdesk reporting metrics. Staffed with over 110 law firm specific helpdesk specialists, Intelliteach currently supports over 80,000 law firm users, handling over 50,000 legal support questions each month. Intelliteach clients are comprised of firms with as few as 25 attorneys to over 2,000+ attorneys including over 30% of the AmLaw 200. With offices in Atlanta, Georgia and London, England, Intelliteach provides support to over 400 law firm locations in 20+ countries around the globe, 24/7. To find out more about Intelliteach or its services please visit their website at.